Tuber Flesh Colour, Enzymatic Discolouration, Dormancy and Late Blight Resistance of 29 Tuber-Bearing Accessions of Solanum spp.

Potato relatives adapted to many different habitats are promising sources of desirable agricultural traits for potato breeding. Wild potato relatives are preserved in different collections around the world, and their detailed description is key to their exploitation in practice. We described 29 seed accessions of 26 Solanum species originating from the VIR potato collection (Institute of Plant Industry – VIR, Saint Petersburg, Russia) and preserved in Poland. The description included resistance to Phytophthora infestans, tuber flesh colour, enzymatic discolouration and tuber dormancy (sprouting). Up to 13 genotypes were evaluated per accession. The evaluation was repeated in three years for each trait. Two P. infestans isolates were used in late blight resistance tests. Amongst all the tested accessions, five were resistant to both P. infestans isolates, including genotype 13_A2. Twenty-one accessions had white tuber flesh, and 13 accessions showed a lack of or weak enzymatic discolouration. Additionally, we found accessions that were whiter than the standard white-fleshed Polish potato cultivar Irys. In our material, we observed a large variation in the length of the sprouts after storage, indicating differences in the dormancy period length. Four accessions showed a lack of sprouting after 28 weeks of storage at 5–6 °C. The tested material is preserved as in vitro plants in the National Centre for Plant Genetic Resources: Polish Genebank (IHAR-PIB, Radzików, Poland), which will facilitate their use in breeding programs.


Introduction
The cultivated potato Solanum tuberosum L. is the third most important food crop in the world after rice and wheat in terms of human consumption (Bethke et al. 2017). In 2019, 370 million tons of potatoes were produced in approximately 150 countries over a total area of 17.3 million hectares (FAOSTAT 2021). It belongs to the genus Solanum, section Petota, which now includes 122 species (Spooner et al. 2014). Wild and cultivated potato relatives originate from America, particularly from Peru, Mexico, Bolivia, Argentina, Venezuela, Colombia and Ecuador. They are adapted to diverse habitats e.g. cloud forests, deserts, scrub vegetation, mountain pastures and volcanic or loamy soils (Machida-Hirano 2015). Wild potato species differ in their ploidy levels (diploids, triploids, tetraploids, hexaploids) and are very diverse in morphological traits such as plant height, leaf and leaflet shape, flower colour, stolon length and size, colour and shape of tubers (Spooner et al. 2004). This diversity makes potato relatives good sources of resistance against a broad range of pests, diseases and abiotic stresses, as well as sources of novel alleles affecting tuber quality and agronomic traits (Hawkes 1994). Wild and cultivated potato species have been used by breeders in potato breeding programs for over 150 years (Bethke et al. 2017). They have been widely collected and maintained as botanical seeds, tubers or in vitro plants in genebanks worldwide. The major collections are in South America, North America, Europe and a few countries in Asia. Within the Cornell-Eastern Europe-Mexico (CEEM) Project on Late Blight Control conducted in 1996-2000 (Raman et al. 2000), 111 accessions of wild and cultivated potato from the VIR collection (Institute of Plant Industry -VIR, Saint Petersburg, Russia) were preserved at the Plant Breeding and Acclimatization Institute -National Research Institute (IHAR-PIB, Radzików, Poland) in the National Centre for Plant Genetic Resources: Polish Genebank (KCRZG). The VIR potato collection was initiated by Bukasov, Voronov and Juzepczuk in 1925-1926and in 1926-1928 by Juzepczuk. They collected and described wild and cultivated potato species discovered in Central and South America (Spooner et al. 2014;Zoteyeva et al. 2012). The goal of this study was to characterise 29 accessions of 26 Solanum species originating from the VIR potato collection in terms of traits such as tuber flesh colour, enzymatic discolouration of tuber flesh, tuber dormancy (sprouting) and foliage resistance to Phytophthora infestans (Mont.) de Bary.
The first trait that potato breeders searched for amongst wild potato species was resistance to P. infestans. This pathogen causes late blight, one of the most economically important potato diseases worldwide. Most potato cultivars are susceptible to late blight; therefore, this disease is managed by intensive chemical control. Yield losses and the cost of fungicides used to control the disease are valued at six billion USD globally (Haverkort et al. 2008). Resistance breeding is an alternative to chemical protection against late blight. The first dominant resistance genes, R1-R11, were identified in the Mexican species S. demissum (Black et al. 1953;Malcolmson and Black 1966). Some of these genes were introgressed by crossing and backcrossing into potato cultivars, such as Pentland Ace with The dormancy of potato tubers is a physiological state after harvest and is useful when the conditions are unfavourable for growth in many regions of the world. Through this resting period, storage of potato tubers becomes possible, and depending on the cultivar, conditions (temperature, humidity and atmospheric composition), geographic location and climate, it may be possible for days, weeks or months (Daniels-Lake and Prange 2007). When the resting period is longer, the tuber quality after storage is better. Earlier sprouting of potato tubers leads to significant economic losses as a result of a loss of water, remobilisation of starch and proteins and shrinkage of tubers (Sonnewald 2001). A commonly used method to inhibit potato sprouting is storage of tubers at low temperatures of 3-7 °C, but this sometimes results in tissue sweetening (Alamar et al. 2017). Chlorpropham (CIPC) has been a commercially used chemical sprout inhibitor for more than four decades (Mohammed et al. 2015;Sorce et al. 2005). Because of its negative effects on human health and the natural environment, many countries restrict or prohibit the use of CIPC (Shukla et al. 2019). The EU Commission published Implementing Regulation (EU) 2019/989 on 17 June 2019, which did not renew approval for the use of this active substance. Research is being carried out to find alternative substances that limit sprouting (Shukla et al. 2019). Tuber dormancy is a complex, polygenic trait influenced by many factors. QTLs for tuber dormancy were found on potato chromosomes II, III, IV, V, VII, VIII and X (Bisognin et al. 2018;Freyre et al. 1994;Sharma et al. 2021).
Potato cultivation suffers from quantitative and qualitative losses caused by biotic and abiotic stresses. Our study enriches the knowledge about useful traits of wild potato relatives preserved in the KCRZG IHAR-PIB. Well-characterised collections of wild potato species can support changing trends in potato breeding e.g. minimising the negative impact on the environment or obtaining potato cultivars with tuber flesh colours attractive to consumers.

Plant Materials
We used 29 seed-preserved accessions of 26 tuber-bearing species stored in the KCRZG IHAR-PIB (Table 1). The selected accessions were from the VIR collection, and the resistance of some of them to P. infestans, potato virus X (PVX) and potato virus Y (PVY) was previously described by Zoteyeva et al. (2012). Four accessions (POL003:333119, POL003:333125, POL003:333112, POL003:333071) have not been tested for resistance to P. infestans before. To promote germination, seeds were soaked in gibberellic acid (700 ppm) for 24 h at room temperature before sowing. After that, 30 seeds per accession were sown in plastic pots in a greenhouse over 2 years: 2016 (15 accessions) and 2017 (14 accessions). From one to 30 seedlings per accession were individually transferred into pots. For each accession, one to 13 well-developed plants were grown in a greenhouse until maturity (Table 1). Then, the tubers were collected separately from each plant and stored at 5-6 °C for further evaluation. In the next years,    Code not available (Huamán and Ross 1985) c Endosperm Balance Number three tubers of each genotype were separately planted in plastic pots. Tubers of the genotypes without sprouts were soaked in thiourea (1%) for 15 min to accelerate sprouting (Pietrak 2001).

Screening for Resistance to P. infestans
Laboratory detached leaflet/leaf assays (DLAs) were performed in 2017-2020 on two different dates and in two replications each year as described by Brylińska and Śliwka (2017). Two Polish isolates of P. infestans MP324 and MP1777 (clonal lineage 13_A2) were used, each isolate in three years (Table 2). Three lateral leaflets (for plants with compound leaves) or one leaf (for plants with simple leaves or with very small, fragile compound leaves from accession numbers POL003:333138, POL003:333155, POL003:333071, POL003:333125) were collected from the middle part of greenhouse-grown, 6-week-old plants from each of the tested genotypes. The leaflets/leaves were placed abaxial side up on wet paper in plastic trays. The inoculum was prepared as described by Sobkowiak and Śliwka (2017). A 30-µl drop of a sporangia suspension with a concentration of 50 sporangia × μl −1 was placed near the midrib on each leaflet. The whole leaves of the four accessions POL003:333138, POL003:333155, POL003:333071, POL003:333125 were sprayed with a sporangia suspension with a concentration of 50 sporangia × μl −1 . The inoculated leaflets and leaves were incubated at 16 °C in the dark and under high-humidity conditions (relative air humidity > 80%). The following day, the leaflets and leaves were turned over, adaxial side up and constant light of approximately 1600 lx was switched on. Scoring was conducted after 6 days of incubation and was performed on a 1-9 scale, where 9 means no disease symptoms (Brylińska and Śliwka 2017). Each test included standard cultivars Craigs Royal, Bzura, Sárpo Mira and Biogold; the tetraploid breeding line 04-IX-21 with the Rpi-phu1 gene; and the diploid breeding lines DG 99-10/36 with the Rpi-rzc1 gene and DG 99-12/8 with the Rpi-mch1 gene, which were described by Janiszewska et al. (2021).

Tuber Flesh Colour and Enzymatic Discolouration
These traits were evaluated 15 weeks after harvest in February 2019, 2020 and March 2021. Tuber flesh colour was determined with two methods. In the first method, flesh colour was scored visually on five tubers per genotype cut in half from the apical to the distal end according to a 1-6 scale, where 1 = white, 2 = grey white, 3 = cream, 4 = light yellow, 5 = yellow, 6 = deep yellow. In the second method, the flesh colour was estimated using a Minolta CR-400 colorimeter (Osaka, Japan), and the yellowness index (YI) was calculated for five tubers per genotype according to the formula: where Cx and Cz are illuminant-and observer-specific constants, and X, Y and Z are trichromatic values (ASTM 2005;Hunter and Harold 1987). Enzymatic discolouration was assessed according to the method described by Hara-Skrzypiec (2017). Five tubers of each genotype were cut in half from the apical to the distal end, and after 4 h, the degree of discolouration was scored according to the colour chart (Dansk Gaerings-Industri, Ltd., Copenhagen, Denmark) on a 1-9 scale, where 9 means a lack of discolouration.
Polish potato cultivars Irys with white flesh colour (1), Harpun with cream flesh colour (3) and Bartek with light yellow flesh colour (4) were used as standards in the assessment of tuber flesh colour and enzymatic discolouration.

Tuber Dormancy (Sprouting)
Sprouting was evaluated 28 weeks after harvest at the end of May 2018, 2019 and 2020. Two weeks before measuring sprouts, the tubers were moved from 5-6 °C to room temperature. One apical potato sprout was measured from 15 tubers of each genotype using electronic callipers CD-15DAX (Mitutoyo Poland Sp. z o.o., Wrocław, Polska).

Data Analyses
STATISTICA for Windows (StatSoft Polska, Kraków, Poland) was used to perform all the statistical analyses. The effects of interactions on the DLAs, flesh colour, enzymatic discolouration and length of sprouts were estimated by Analysis of variance (ANOVA). In the DLAs, the years of testing for the two pathogen isolates were different (Table 2). Therefore, for the DLAs in ANOVA, we changed from a sigmarestricted model to a type IV sum-of-squares model. Duncan's multiple range test was used to assess the significance of differences in the DLAs between the standard cultivars. The correlations between the results of resistance to two P. infestans isolates, MP324 and MP1777, and between the results of the two methods of flesh colour assessment were determined through calculation of Pearson's correlation

Resistance to P. infestans
The 29 accessions of 26 Solanum species were tested for their resistance against two P. infestans isolates, MP1777 (clonal lineage 13_A2) and MP324, as described in Table 2. The tests with each isolate were repeated in three different years. Phytophthora infestans isolate MP1777 was virulent towards Black's differential plants with the R5 and R10 genes ( Table 2) and more aggressive than isolate MP324, which is illustrated by the reaction of standard cultivars and breeding lines (Fig. 1). The 3-year mean resistance scores of standards were lower for isolate MP1777 than for isolate MP324 and amounted to 2.9-8.8 and 4.1-9.0, respectively. A significant difference in P. infestans resistance scores for both isolates was observed for the Biogold cultivar, with mean values of 5.0 (MP1777) and 7.3 (MP324). The 3-year mean resistance scores in the DLAs obtained with both isolates were strongly correlated   (Table 3). Accession had the largest influence on resistance to P. infestans. For 22 accessions, the results obtained with the two P. infestans isolates did not differ from each other. Plants of all tested genotypes from five accessions were resistant to P. infestans isolates MP324 and MP1777, with 3-year mean resistance scores from 9 to 6 on a 1-9 scale ( Fig. 2; Table 4). Segregation of the resistance to both isolates was noted amongst plants from eight accessions. For nine accessions, 3-year mean scores indicated that all plants were susceptible to P. infestans isolates. Differences in the reaction to P. infestans isolates MP324 and MP1777 were noted in seven accessions. All plants of accession POL003:333159 of the species S. papita were resistant to isolate MP324 but were segregated in terms of their  Fig. 2A, B; Table 4).

Tuber Flesh Colour and Enzymatic Discolouration
Flesh colour was not assessed by a visual method for one accession POL003:333071 of species S. uyunense due to the low number of tubers. Colorimeter measurements were not possible due to the small size of the tubers for four accessions (S. antipovichii POL003:333099, S. punae POL003:333138, S. uyunense POL003:333071, S. acaule POL003:333155). ANOVA revealed the main effect of accession on flesh colour determined by a visual method and the main effect of accession and year on this trait assessed by a colorimeter (YI) ( Table 3). The results obtained by both methods were highly correlated (the Pearson's correlation coefficient between the 3-year mean results was 0.855 at p < 0.001). Twenty-one accessions had white (1) to cream (3) flesh colour determined by the visual method (Fig. 3A). Amongst seven accessions, segregation from white (1) to yellow (5) flesh colour was observed. No anthocyanin-derived colours were observed in the tested accessions. The yellowness index of the assessed wild potato species was characteristic of white-fleshed potato for 20 accessions, and the maximum value of the 3-year mean was 48.5 (Fig. 3B). Segregation of the 3-year mean yellowness index from whitefleshed to yellow-fleshed potato was observed amongst five accessions. The standard cultivars had the following 3-year mean yellowness index scores: Irys, 40.5; Harpun, Consistent results from both methods of tuber flesh colour assessment were observed for 23 accessions. For the two accession numbers there were differences between the methods. The flesh colour of accessions POL003:333121 of species S. kurtzianum and POL003:333149 of species S. microdontum determined by visual method segregated between white and light-yellow flesh, but the yellowness index indicated that they all had white flesh (Fig. 3). Mean enzymatic discolouration of the tested potato accessions evaluated in 2019-2021 is shown in Fig. 3. ANOVA demonstrated significant effects of accession, year and the accession × year interaction on enzymatic discolouration (Table 3). Thirteen accessions showed a lack of or weak enzymatic discolouration, with 3-year mean values of 7-9 on a 1-9 scale. Segregation of mean enzymatic discolouration scores from 9 to 4 was observed for 15 accessions. Accession POL003:333069 of the species S. parodii developed the strongest enzymatic discolouration, with a 3-year minimum of 5.5 (Fig. 4). Standard cultivars Irys, Harpun and Bartek showed weak enzymatic discolouration, with a 3-year mean of 8.

Tuber Dormancy (Sprouting)
ANOVA showed significant effects of accession, year and the accession × year interaction on sprouting (Table 3). Four accessions (S. papita POL003:333147, S. fendleri POL003:333110, POL003:333112, S. aemulans POL003:333119) showed a lack of sprouting after 28 weeks of storage (Fig. 5). Sprouts up to 10 mm (3-year mean) were observed for eight accessions. The mean length of sprouts was segregated for the remaining 17 accessions. The longest sprouts (98.3 mm on average) were described for plants of the accession POL003:333141 of species S. simplicifolium.

Discussion
Amongst the most important crops in the world, potato has the highest number of wild relatives (Vincent et al. 2013). Screening of collections of wild and cultivated potato species, which are preserved in different gene banks, is a time consuming and laborious process. The potato genetic resources preserved as seeds are often disordered and unexplored, with variation amongst individuals within and across accessions and with some species overrepresented and others not adequately represented. Storage of accessions as seeds makes it difficult for breeders to utilise these reservoirs of valuable alleles. To date, characteristics of germplasm collections have been based on populations rather than individuals, without linking individual genotypes to phenotype and genotype data (Bethke et al. 2017). Pérez et al. (2001) tested the late blight resistance of 139 population samples of 51 Solanum species, which were preserved as seeds in the International Potato Center (CIP, Lima, Peru). They found all susceptible plants for 22 accessions, all resistant plants for 7 accessions and segregation of resistance for 110 accessions. One hundred ninety-eight accessions of 63 different Solanum species from The Commonwealth Potato Collection (CPC, Invergowrie, Great Britain) were mostly preserved as seeds, and few as tubers were examined for resistance to Globodera pallida and G. rostochiensis by Castelli et al. (2003). They identified 56% and 53% clones resistant to G. pallida and G. rostochiensis, respectively, amongst all tested accessions. Khiutti et al. (2015) tested up to three and to the viruses PVX and PVY. They evaluated the resistance of accessions to two P. infestans isolates, MP322 in 1998and MP324 in 1999. The results of this work have been exploited to only a limited extent, as two major resistance genes against P. infestans have been identified and mapped in this material: Rpi-rzc1 from S. ruiz-ceballosii (Śliwka et al. 2012a) and Rpi-mch1 from S. michoacanum (Śliwka et al. 2012b). The Rpi-rzc1 gene has been transferred to the tetraploid level of cultivated potato and pyramided on diploid and tetraploid levels with the Rpi-phu1 gene from S. phureja (Stefańczyk et al. 2020). Smyda-Dajmund et al. (2017) introgressed the Rpi-mch1 gene into the cultivated potato gene pool. One of the reasons for the limited use of material from the work of Zoteyeva et al. (2012) was the loss of clonally propagated and characterised wild potato genotypes. The individual genotypes from our work were preserved as in vitro plants and are available in KCRZG IHAR-PIB, which should have a positive effect on the use of these resources. Smyda-Dajmund et al. (2020) also evaluated the cytoplasm types for all the genotypes of accessions from our work, which will allow more effective exploitation in breeding programs. From the accessions that were tested earlier by Zoteyeva et al. (2012) Emerging new strains of the pathogen, for example, by sexual recombination, can change frequently and easily be transported along with wind-borne spores and seed tubers between individual countries. Therefore, it is important to search for new sources of resistance against P. infestans. In this study, we identified five accessions (S. guerreroense POL003:333096, S. neoantipovichii POL003:333117, S. papita POL003:333147, S. microdontum POL003:333149, and S. antipovichii POL003:333099) that were highly or moderately resistant to both P. infestans isolates. Our results on resistance to P. infestans isolate MP324 for 17 accessions were consistent with those reported earlier by Zoteyeva et al. (2012). Two accessions (POL003:333096 of species S. guerreroense and POL003:333069 of species S. parodii) were not tested with isolate MP324 by these authors. Moreover, four accessions (S. papita POL003:333159 and POL003:333147, S. microdontum POL003:333149, S. polytrichon POL003:333108) were resistant in our research, but in Zoteyeva et al. (2012), segregation of resistance was observed. In addition, amongst two accessions (S. hougasii POL003:333148, S. famatinae POL003:333139) resistance segregated in this work, but both were susceptible in the previous study. The observed differences between our results and those of Zoteyeva et al. (2012) are due to the loss of virulence of the MP324 isolate towards Black's differential plants with the R5, R8 and R10 genes.
Our results indicated the presence of new alleles related to a white flesh colour amongst 29 tested accessions of 26 Solanum species because both a visual method of flesh colour assessment and using a colorimeter revealed genotypes that were whiter than the standard cultivar Irys, described in the cultivar catalogue as having a white flesh colour. Amongst the tested accessions, we identified those that had a white flesh colour and no undesirable browning, which can be desirable traits for breeders (S. stoloniferum POL003:333100, S. guerreroense POL003:333096, S. papita POL003:333147). We also observed large variation in sprouting amongst the accessions and even within genotypes in the same accession. We noticed accessions lacking sprouts after 28 weeks of storage (S. papita POL003:333147, S. fendleri POL003:333110, POL003:333112, S. aemulans POL003:333119), which can contribute to the identification of new alleles and be used to obtain potato cultivars suitable for long-term storage. Bamberg (2010) even described the occurrence of an 8-year tuber dormancy period in S. jamesii. This study provides information on approximately 29 accessions of 26 Solanum species originating from the VIR potato collection and preserved in KCRZG IHAR-PIB. Some of these accessions could be an important source of traits related to resistance to P. infestans, reduced or no enzymatic discolouration in tubers, a white flesh colour and a long tuber dormancy period.